Showing papers presented at "International Conference on Distributed Computing Systems Workshops in 2022"

Life, the Metaverse and Everything: An Overview of Privacy, Ethics, and Governance in Metaverse

Abstract: The metaverse is expected to be the next major evolution phase of the internet. The metaverse will impact human society, production, and life. In this work, we analyze the current trends and challenges that building such a virtual environment will face. We focus on three major pillars to guide the development of the metaverse: privacy, governance, and ethical design, to guide the development of the metaverse. Finally, we propose a preliminary modular-based framework for an ethical design of the metaverse.

Re-shaping Post-COVID-19 Teaching and Learning: A Blueprint of Virtual-Physical Blended Classrooms in the Metaverse Era

Abstract: During the COVID-19 pandemic, most countries have experienced some form of remote education through video conferencing software platforms. However, these software platforms fail to reduce immersion and replicate the classroom experience. The currently emerging Metaverse addresses many of such limitations by offering blended physical-digital environments. This paper aims to assess how the Metaverse can support and improve e-learning. We first survey the latest applications of blended environments in education and highlight the primary challenges and opportunities. Accordingly, we derive our proposal for a virtual-physical blended classroom configuration that brings students and teachers into a shared educational Metaverse. We focus on the system architecture of the Metaverse classroom to achieve real-time synchronization of a large number of participants and activities across physical (mixed reality classrooms) and virtual (remote VR platform) learning spaces. Our proposal attempts to transform the traditional physical classroom into virtual-physical cyberspace as a new social network of learners and educators connected at an unprecedented scale.

Social games and Blockchain: exploring the Metaverse of Decentraland

Abstract: Online Social Networks gained a crucial role in people's everyday life, acting as the medium through which people can interact with each other. The introduction of blockchain technology prompted a new generation of social media based on the concept of Non-Fungible Token (NFT) and other Web3 technologies, giving birth to the Metaverse. Decentraland is one such platform, where users can explore a 3D virtual world and communicate with each other. The main feature of Decentraland is that the virtual world is divided into parcels, implemented through NFTs, that can be traded among users and where owners can create buildings or mini-games for other users to play with. In this work, we explore the virtual world of Decentraland by presenting the main details of the virtual world and by focusing on the economic impact of NFT trading on the description of the parcels. In detail, the parcel's description can be used to advertise the proximity of the parcel to infrastructures or special projects to attract potential buyers. This paper's findings show the impact and the magnitude of this phenomenon, highlighting entire quarters that adopt this technique throughout the map by overriding the playful aspect of the platform.

Passively Measuring IPFS Churn and Network Size

Abstract: The InterPlanetary File System (IPFS) is a popular decentralized peer-to-peer network for exchanging data. While there are many use cases for IPFS, the success of these use cases depends on the network. In this paper, we provide a passive measurement study of the IPFS network, investigating peer dynamics and curiosities of the network. With the help of our measurement, we estimate the network size and confirm the results of previous active measurement studies.

Fair Incentivization of Bandwidth Sharing in Decentralized Storage Networks

Abstract: Peer-to-peer (p2p) networks are not independent of their peers, and the network efficiency depends on peers contributing resources. Because shared resources are not free, this contribution must be rewarded. Peers across the network may share computation power, storage capacity, and bandwidth. This paper looks at how bandwidth incentive encourages peers to share bandwidth and rewards them for their contribution. With the advent of blockchain technology, many p2p networks attempt to reward contributions by crypto-assets. We conduct simulations to better understand current incentive mechanisms, assess the fairness of these mechanisms, and to look for ways to make those incentives more equitable. The following are the primary contributions of this study: (i) We investigate and simulate bandwidth incentives within Swarm, a cutting-edge p2p storage network; (ii) We demonstrate one approach to make the current bandwidth incentives more equitable; (iii) We use the Gini coefficient to define two quantifiable fairness characteristics to evaluate reward sharing in a decentralized p2p storage network.

Enriching Kademlia by Partitioning

Abstract: Decentralizing the Web is becoming an increasingly interesting endeavor that aims at improving user security and privacy as well as providing guaranteed ownership of content. One such endeavor that pushes towards this reality, is Protocol Labs' Inter-Planetary File System (IPFS) network, that provides a decentralized large scale file system to support the decentralized Web. To achieve this, the IPFS network leverages the Kademlia DHT to route and store pointers to content stored by network members (i.e., peers). However, due to the large number of network peers, content, and accesses, the DHT routing needs to be efficient and quick to enable a decentralized web that is competitive. In this paper, we present work in progress that aims at improving the Kademlia DHT performance through the manipulation of DHT identifiers by adding prefixes to identifiers. With this, we are able to bias the DHT topological organization towards locality (which can be either geographical or applicational), which creates partitions in the DHT and enables faster and more efficient query resolution on local content. We designed prototypes that implement our proposal, and performed a first evaluation of our work in an emulated network testbed composed of 5000 nodes. Our results show that our proposal can benefit the DHT look up on data with locality with minimal overhead.

Hierarchical Consensus: A Horizontal Scaling Framework for Blockchains

Abstract: We present the Filecoin Hierarchical Consensus framework, which aims to overcome the throughput challenges of blockchain consensus by horizontally scaling the network. Unlike traditional sharding designs, based on partitioning the state of the network, our solution centers on the concept of subnets -which are organized hierarchically- and can be spawned on-demand to manage new state. Child sub nets are firewalled from parent subnets, have their own specific policies, and run a different consensus algorithm, increasing the network capacity and enabling new applications. Moreover, they benefit from the security of parent subnets by periodically checkpointing state. In this paper, we introduce the overall system architecture, our detailed designs for cross-net transaction handling, and the open questions that we are still exploring.

Edge computing system with multi-LIDAR sensor network for robustness of autonomous personal-mobility

Abstract: One of the most important systems for smart cities is an edge computing system to facilitate decision-making by using real-time awareness and data analytics. Although existing works focused on the safety of pedestrians, the safety of personal- mobility vehicles is what smart cities need to address toward the future. On-board sensors could be easily disabled because of external issues of hardware caused by dirt or snow, internal issues of hardware, and software issues such as malware infection. We propose an edge computing system that supports robust vehicle automation with a sensor network using multiple light-detection- and-ranging (LIDAR) sensors deployed like roadside cameras or indoor security cameras for surveillance for personal-mobility vehicles. We develop a prototype system of the proposed system using real LIDAR units and a mobility scooter and evaluate the system in terms of route errors to verify the feasibility of the proposed system with a scenario where the vehicle is enabled to continue self-driving even when its on-board sensor is disabled.

Towards an Open Source Fully Modular Multi Unmanned Aerial Vehicle Simulation Framework

Abstract: With the trend towards smart cities, the use of Unmanned Aerial Vehicles (UAVs) is becoming more and more diverse. An economical and insightful way to study the use of UAV s in cities is through the utilization of computer simulations. However, currently available simulators are only of limited use to investigate such scenarios: often they have been developed for a dedicated use case, can only handle single UAV s, or the software was not made available with an open license. In this paper, we present AirMobiSim, a modular and microscopic UAV simulation framework, which is available under a GPLv2 license. AirMobiSim provides the basis for the creation of kinematic and energy models for different UAV types. It can support an arbitrary number of UAV s and can be coupled with other simulators via open interfaces, for example to examine systems using wireless communication between UAVs. We furthermore present our approach for model-building and show that AirMobiSim can already accurately reproduce existing work. We also highlight current limitations and show prospects for future work.

Cost-Effective Optimal Multi-Source Energy Management Technique in Heterogeneous Networks

Abstract: Heterogeneous networks (HetNets) involve multi-source energy supply for energy consumption through energy conservation and emission reduction. This research paper considers traditional energy's Torono Real-Time (TOU) price and fairness among users. It uses exponential utility functions that build a relationship between energy consumption, energy cost, and user fairness. Furthermore, it divides user selection into two stages such as (i) edge user allocation and (ii) service user determination to obtain an optimally balanced load distribution. We evaluate the proposed technique by combining two algorithms such as PA1 and PA2. The proposed approach significantly reduces the cost and energy consumption of the base station system in heterogeneous networks.

Decentralized Hole Punching

Abstract: We present a decentralized hole punching mechanism built into the peer-to-peer networking library libp2p [1]. Hole punching is cru-cial for peer-to-peer networks, enabling each participant to directly communicate to any other participant, despite being separated by firewalls and NATs. The decentralized libp2p hole punching proto-col leverages protocols similar to STUN (RFC 8489 [2]), TURN (RFC 8566 [3]) and ICE (RFC 8445 [4]), without the need for any cen-tralized infrastructure. Specifically, it doesn't require any previous knowledge about network participants other than at least one (any arbitrary) node to bootstrap peer discovery. The key insight is that the protocols used for hole punching, namely address discovery and relaying protocols, can be built such that their resource require-ments are negligible. This makes it feasible for any participant in the network to run these, thereby enabling the coordination of hole punch attempts, assuming that at least a small fraction of nodes is not located behind a firewall or a NAT.

Deep Learning Based Joint Collision Detection and Spreading Factor Allocation in LoRaWAN

Abstract: Long-range wide area network (LoRaWAN) is a promising low-power network standard that allows for long-distance wireless communication with great power saving. $L$ oRa is based on pure ALOHA protocol for channel access, which causes collisions for the transmitted packets. The collisions may occur in two scenarios, namely the intra-spreading factor (intra-SF) and the inter-spreading factor (inter-SF) interference. Consequently, the SFs assignment is a very critical task for the network performance. This paper investigates a smart SFs assignment technique to reduce collisions probability and improve the network performance. In this work, we exploit different architectures of artificial neural networks for detecting collisions and selecting the optimal SF. The results show that the investigated technique achieves a higher prediction accuracy than traditional machine learning algorithms and enhances the energy consumption of the network.

Combining Distributed and Central Control for Self-Adaptive Systems of Systems

Abstract: Distributed and central control are two complementary paradigms to establish self-adaptation in software systems. Both approaches have their individual benefits and drawbacks, which lead to significant trade-offs regarding certain software qualities when designing such systems. The significance of these trade-offs even increases the more complex the target system becomes. In this paper, we present our work-in-progress towards an integrated control approach, which aims at providing the best of both control paradigms. We present the basic concepts of this multi-paradigm approach and outline its inherent support for complex system hierarchies. Further, we illustrate the vision of our approach using application scenarios from the smart energy grid as an example for self-adaptive systems of systems.

Decentralized Technology in Practice: Social and technical resilience in IPFS

Abstract: The “Interplanetary File System” (IPFS) refers to a peer-to-peer content addressing protocol that is designed to be “upgradeable, resilient, and more open”. [1] In conjunction with “Filecoin”, a blockchain-based data storage marketplace, the aim of this technology is to offer people digital infrastructure that decentralized from central, corporate control, to store and manage their data. This paper investigates socio-technical resilience as adaptability and transformability under threat for participants in IPFS in practice as a peer-to-peer digital infrastructure. I find that developer conceptions of “resilience” in this system relate to adaptability of the network against isolation or split. Yet, resilience for those using IPFS relates to data integrity, in line with the preferences and local context of the provider of that data. I employ qualitative research methods to investigate why and how people use the network, and in what ways it meets the needs of their contexts and purposes, or in some cases, falls short. I find that while significant emphasis is placed on the technical resilience of this infrastructure, these tools remain logically fractured at the social coordination layer that guides processes of data ownership, governance, and storage, and this limits the resilience of the network to serve people's needs in their local context. I then propose examples of institutional infrastructural frameworks that could provide participatory data governance patterns to help people collectively organize their use of decentralized digital infrastructure for greater resilience in IPFS and Filecoin. Finally, I outline further research directions to improve resilience in decentralized data storage infrastructure.

Hyperverse: A High Throughput Pattern Matching Engine for Metaverse

Abstract: Cyberspace has continued to evolve since the In-ternet became widespread in the 1990s. A variety of computer-mediated virtual environments have been created, including social networks, video conferencing, virtual 3D worlds (e.g., VR chat), augmented reality applications (e.g., Ingress), and non-fungible token games. Such virtual environments, while not permanent and incoherent, have brought us varying degrees of digital transformation. The term “metaverse” was devised to further facilitate the digital transformation of all aspects of our physical lives. With the increasing number of devices and users such as loT, augmented reality and virtual reality glass connected to the metaverse, huge amounts of data need to be filtered or captured for metaverse security or user behavior analysis by utilizing pattern matching. However, directly utilizing the existing pattern matching engine is impossible since it cannot achieve the throughput that is required by the metaverse, where low throughput could cause poor user experiences. Thus, in this paper, we propose a new pattern matching design called Hyperverse. Hyperverse can significantly increase the throughput of pattern matching by designing a new algorithm that is based on instruction-level parallelism. We implement Hyperverse in Hyperscan, which is the fastest regular expression engine in the world. Compared with the existing solution, Hyperverse can achieve a throughput of up to 10.4Gbps per core, which is a 3.83x boost than the existing solution. Thus, the significantly increased throughput will prevent a negative impact on the user experience in the metaverse.

DeepIIoT: An Explainable Deep Learning Based Intrusion Detection System for Industrial IOT

Abstract: IoT adoption is becoming widespread in different areas of applications in our daily lives. The increased reliance on IoT devices has made them a worthy target for attackers. With malicious actors targeting water treatment facilities, power grids, and power nuclear reactors, industrial IoT poses a much higher risk in comparison to other IoT application contexts. In this pa-per, we present a deep-learning based intrusion detection system for industrial IoT. The proposed system was trained and tested using the WUSTL-IIOT-2021 dataset. Testing results showed accuracy exceeding 99% with minimally low false-positive, and false-negative rates. The proposed model was explained using SHAP values.

Incentive-driven Computation Offloading and Resource Allocation in Mobile Cloud-Edge Computing

Abstract: This paper proposes a novel Reverse Auction-based Computation Offloading and Resource Allocation Mechanism, named RACORAM for the mobile Cloud-Edge computing. The basic idea is that the Cloud Service Center (CSC) recruits edge server owners to replace it to accommodate offloaded computation from nearby resource-constraint Mobile Devices (MDs). In RACORAM, the reverse auction is used to stimulate edge server owners to participate in the offloading process, and the reverse auction-based computation offloading and resource allocation problem is formulated as a Mixed Integer Nonlinear Programming (MINLP) problem, aiming to minimize the cost of the CSC. Specifically, a Greedy Randomized Adaptive Search Procedure based Winning Bid Scheduling Method (GWBSM) is proposed to determine the computation offloading strategy. Simulations are conducted to evaluate the performance of RACORAM, and the results show that RACORAM is very close to the optimal method with significantly reduced computational complexity, and greatly outperforms the other baseline methods in terms of the CSC's cost under different scenarios.

Solving Source Location Problems with Particle Swarm Optimizer and Height Information

Abstract: Recently, there has been great interest in utilizing robot swarm to solve source location problems. A particle swarm optimizer (PSO) is among the most popular method. To our best knowledge, almost all work in this field focuses on a 2-D search space while ignoring height information. This work proposes a 3-D source location model where the sensing range of an unmanned aerial vehicle (UAV) gradually increases with its flying height, but its obtained signal strength gradually weakens. It can avoid meaningless wandering of UAV s in no-signal areas. Experimental results show that the effective use of height information can significantly improve the search efficiency.

Towards Portable Identities in the Matrix Protocol

Abstract: In this paper we present a cryptographic challenge-response authentication mechanism to enable portable identities within the Matrix protocol. Online instant communication systems like Matrix connect people in a convenient and cost-effective manner. However, most of today's communication infrastructure relies on inherently centralized infrastructure. Matrix currently, takes a federated approach instead. However, to fully decentralize the Matrix network, further actions must take place. Among others, a user's identity must be disconnected from the server they used to sign up and happen to communicate on. As part of this effort, a challenge-response authentication mechanism has been implemented that allows user IDs to be usable on any server, even if the a user's homeserver becomes unreachable.

Downscaling and Overflow-aware Model Compression for Efficient Vision Processors

Abstract: Network pruning and quantization are two effective ways for model compression. However, existing model compression methods seldom take hardware into consideration, resulting in compressed models that still take high energy and chip area cost on a vision processor. To address this issue, one may reduce the bit-widths of the accumulator and the multiplier in fixed-point inference to significantly reduce the energy and chip area. However, the numerical error brought from the low-bit multiplier in the downscaling procedure is large, while the low-bit accumulator suffers from the overflow issue. Both of them lead to significant performance degradation. In this paper, we propose downscaling and overflow-aware model compression for efficient vision processors. Specifically, we propose downscaling-aware training to simulate the downscaling procedure during training so that the models are adjusted to inference with low bit-width multipliers. To address the overflow issue, we apply overflow-aware training to gradually reduce the range of quantized values. We further restrict the channel's number of each layer to be the multiple of some value (e.g., 16) to take advantage of parallel computing by channel pruning. With the proposed method, we are able to obtain the compressed model with low bit-width accumulators and multipliers during inference while maintaining the performance. As a result, the energy and chip area cost can be significantly reduced. To demonstrate this, we further co-design an agilely customizable vision processor and its SoC. Extensive experiments on image classification, object detection, and semantic segmentation demonstrate the effectiveness of our proposed method. For example, on ImageNet, our compressed 8-bit ResNet-50 achieves lossless performance with 16-bit accumulators and 12-bit multipliers.

Decentralizing Watchtowers for Payment Channels using IPFS

Abstract: Payment channels have been proposed as a way to improve the scalability of blockchains such as Bitcoin. However, payment channel protocols require that participating parties watch the blockchain regularly for new transactions. If a party observes, in a given period of time, a fraudulent transaction that closes the payment channel in an outdated state, the fraudulent transaction can be revoked. Previous work has proposed to outsource this task to a third party, a so called watchtower. A user of a payment channel employs a dedicated watchtower and sends the data to the watchtower that the watchtower requires to revoke fraudulent transactions. In this paper, we replace the strict binding of a user to a watchtower by a decentralized approach for watchtowers that requires no direct interaction between a party of a payment channel and the watchtower. This decentralized approach uses IPFS to publicly store the information required by a watchtower. With this approach, anyone can detect and revoke a fraud by watching the blockchain and reading a file from IPFS that contains information for each outdated commitment transaction. A reward for successful revocations can be used as incentive.

Planning Computation Offloading on Shared Edge Infrastructure for Multiple Drones

Abstract: Drones are used in a wide range of applications, which may involve computationally-demanding data processing tasks during the missions. While such heavy tasks can be offloaded to nearby edge-servers, this may not always be feasible due to capacity limitations and contention. In this case, it is important to have a fair allocation of server resources to drones. We propose a heuristic for this problem, and evaluate it though simulation experiments using realistic performance parameters. We show that the mission time can be greatly reduced, by up to 33% (16 min) compared to the default where drones perform all computations onboard, while evenly balancing the benefits of offloading among drones with different missions.

Decentralized Self-Adaption With Epidemic Algorithms for Agent-Based Transportation

Abstract: We investigate epidemic algorithms to enable agent-based transporters to be self-adaptive to disturbances and avoid centralized communication mechanisms. We conduct simulation experiments of a shop floor to compare the ability of mobile agents with limited perception to deliver items and adapt to a randomly disturbed environment by communicating decentralized with epidemic algorithms, and centralized via blackboard and direct messages. For evaluation of their adaption, we measure their task performance and communication efficiency. We conclude that agent-based transportation with epidemic algorithms can self-adapt to a disturbed environment, can still perform close to centralized ones, and avoid monolithic components.

A Blockchain-based Profiling System for Exploring Human Factors in Cloud-Edge-End Orchestration

Abstract: In mobile edge computing (MEC), application partitioning is one of the most effective measures to leverage computing resources. Due to the user's unpredictable behavior pattern, which is an indispensable factor affecting the performance of an offloading system, traditional partitioning algorithms, considering only purely technical QoS, are no longer enough to meet the increasing concern for the user experience of mobile applications. In this paper, in order to explore human factors in modeling partitioning algorithms for cloud-edge-end orchestration under a safe and trusted environment, we present a blockchain-based profiling system to collect behavioral data from several invited subjects. For discovering user-driven relations of method-level components, we propose a clustering algorithm framework to process each subject's data. Based on the disparate results, we illustrate a case study to prove the usefulness of the system and the data for the orchestration by analyzing the variance of user behavior and the feasibility of applying human factors to the partitioning algorithm.

Privacy-preserving Blockchain-based Global Data Sharing for Federated Learning with Non-IID Data

Abstract: Federated learning is a popular privacy-enhanced distributed machine learning method that solves the problem of local data privacy by gathering the training results (such as model weights, gradients, etc.) instead of the raw data to generate a global model. But a practical problem it faces is the non-independent and identical distribution of data, which means the local data of each participant is highly inconsistent, both in terms of quantity and distribution. Moreover, there is a lack of research related to the efficiency and privacy issues in the pre-training process. Therefore, in this paper, we propose a novel solution that utilizes blockchain technology to realize small-scale global data sharing to assist the training progress. Simulation experiments verify that our method not only guarantees data security but also greatly improves performance in terms of training speed and accuracy.

Inducing Defenders to Mislead an Attacking UAV Swarm

Abstract: In recent years, the development of self-organizing and autonomous behaviors for unmanned aerial vehicle (UAV) swarms has increased significantly. Being flexible, scalable and robust, UAV swarms bring many advantages for future applications. However, these properties might also be used for malicious or dangerous applications like autonomous target-oriented attacks. To date, defense includes strategies like fighting the attackers with a defender swarm or exploiting hardware devices like nets and jammers to stop the attackers. These solutions increase the risk on collateral damage even further. To the best of our knowledge, research is lacking intelligent countermeasures against attacking UAV swarms which limit the damage as much as possible. In this paper, we explore how to invert a robust UAV swarm behavior by inducing defender UAVs into an attacking UAV swarm with the goal to mislead the swarm's mission. Via simulations, we model two different swarm behaviors and explore how to invert them with disguised UAVs deflecting the entire swarm.

Towards Efficient Decentralized Federated Learning

Abstract: We focus on the problem of efficiently deploying a federated learning training task in a decentralized setting with multiple aggregators. To that end, we introduce a number of improvements and modifications to the recently proposed IPLS protocol. In particular, we relax its assumption for di-rect communication across participants, using instead indirect communication over a decentralized storage system, effectively turning it into a partially asynchronous protocol. Moreover, we secure it against malicious aggregators (that drop or alter data) by relying on homomorphic cryptographic commitments for efficient verification of aggregation. We implement the modified IPLS protocol and report on its performance and potential bottlenecks. Finally, we identify important next steps for this line of research.

On the BER Performance of RIS-Enhanced NOMA-Assisted Backscatter Communication under Nakagami-m Fading

Abstract: Backscatter communication (BackCom) has been envisioned as a prospective candidate for enabling the sustained operation of battery-constrained Internet-of- Things (IoT) devices. This approach involves the transmission of information by a backscatter node (BSN) through passive reflection and modulation of an impinging radio-frequency (RF) signal. However, the short operational range and low data rates of contemporary BackCom systems render them insufficient on their own to provide ubiquitous connectivity among the plethora of IoT devices. Meanwhile, wireless networks are rapidly evolving towards the smart radio paradigm. Thus, to enhance the coverage range and capacity, reconfigurable intelligent surfaces (RISs) can be incorporated into the existing BackCom systems. RISs employ passive reflective elements to adaptively configure the stochastic wireless environment in a cost-effective and energy- efficient manner. Furthermore, non-orthogonal multiple access (NOMA) can be exploited to improve the spectral efficiency of the BackCom systems. In this paper, we present the design and bit error rate (BER) analysis of an RIS-enhanced NOMA-assisted bistatic BackCom system under Nakagami-m fading channel. Our extensive simulation results reveal the effectiveness of the proposed system over the conventional NOMA-assisted BackCom system without RIS, and demonstrate the impact of various factors, including the power-reflection coefficients, RIS phase-shift designs, number of reflecting elements, RIS location, and split factor, on the BER performance of the proposed RIS-assisted system.

QUIC Employment: Comparing the Response Time of Facebook and Twitter

Abstract: TCP has been the most reliable transport layer protocol, used on the Internet since 1974, thanks to its ordered and error-checked delivery, stateful property and retransmission property. The most relevant drawback is the high latency, meaning a connection that does not require a reliable delivery can use UDP, another transport layer protocol, slimmer than TCP, which provides a connectionless datagram service prioritizing time over reliability. During these years, researchers have worked on new protocols in the attempt of decreasing page load time without sacrificing reliability. Even Google proposed a solution, the Quick UDP Internet Connection (QUIC) protocol, born in 2013. Since QUIC is a very recent protocol, this paper could help to further understanding its employment and performance in typical scenarios such as, for instance, when considering major social networks such as Facebook and Twitter, with only the former employing QUIC. In particular, we have measured the response time of the two applications when considering different network conditions. Our preliminary results could be used to generate traffic models and, partially, to advocate the use of QUIC.

Waku: A Family of Modular P2P Protocols For Secure & Censorship-Resistant Communication

Abstract: Waku is a family of modular protocols that enable secure, censorship-resistant, and anonymous peer-to-peer communication. Waku protocols provide capabilities that make them suitable to run in resource-restricted environments e.g., mobile devices and web browsers. Such capabilities include (i) retrieving historical messaging for mostly-offline devices (ii) adaptive nodes; allowing for heterogeneous nodes to contribute to the network (iii) preserving bandwidth usage for resource-restricted devices, (iv) minimizing connectivity requirements for devices with a limited connection, and (v) enabling efficient, private, economic spam protection for heterogeneous nodes. Waku's modular design and resource-efficient protocols make it superior to its predecessor i.e., Whisper. In this paper, we give an overview of the Waku protocols stack, its architecture, and protocols interaction along with a sample demo scenario on configuring and running a Waku node using nwaku i.e., Waku client written in Nim.