Mobile ad hoc routing protocols allow nodes with wireless adaptors to communicate with one another without any pre-existing network infrastructure. Existing ad hoc routing protocols, while robust to rapidly changing network topology, assume the presence of a connected path from source to destination. Given power limitations, the advent of short-range wireless networks, and the wide physical conditions over which ad hoc networks must be deployed, in some scenarios it is likely that this assumption is invalid. In this work, we develop techniques to deliver messages in the case where there is never a connected path from source to destination or when a network partition exists at the time a message is originated. To this end, we introduce Epidemic Routing, where random pair-wise exchanges of messages among mobile hosts ensure eventual message delivery. The goals of Epidemic Routing are to: i) maximize message delivery rate, ii) minimize message latency, and iii) minimize the total resources consumed in message delivery. Through an implementation in the Monarch simulator, we show that Epidemic Routing achieves eventual delivery of 100% of messages with reasonable aggregate resource consumption in a number of interesting scenarios.

/pdf/epidemic-routing-for-partially-connected-ad-hoc-networks-s5rslcaiip.pdf

Epidemic routing for partially-connected ad hoc networks

For small cell technology to significantly increase the capacity of tower-based cellular networks, mobile users will need to be actively pushed onto the more lightly loaded tiers (corresponding to, e.g., pico and femtocells), even if they offer a lower instantaneous SINR than the macrocell base station (BS). Optimizing a function of the long-term rate for each user requires (in general) a massive utility maximization problem over all the SINRs and BS loads. On the other hand, an actual implementation will likely resort to a simple biasing approach where a BS in tier j is treated as having its SINR multiplied by a factor Aj ≥ 1, which makes it appear more attractive than the heavily-loaded macrocell. This paper bridges the gap between these approaches through several physical relaxations of the network-wide association problem, whose solution is NP hard. We provide a low-complexity distributed algorithm that converges to a near-optimal solution with a theoretical performance guarantee, and we observe that simple per-tier biasing loses surprisingly little, if the bias values Aj are chosen carefully. Numerical results show a large (3.5x) throughput gain for cell-edge users and a 2x rate gain for median users relative to a maximizing received power association.

User Association for Load Balancing in Heterogeneous Cellular Networks

For small cell technology to significantly increase the capacity of tower-based cellular networks, mobile users will need to be actively pushed onto the more lightly loaded tiers (corresponding to, e.g., pico and femtocells), even if they offer a lower instantaneous SINR than the macrocell base station (BS). Optimizing a function of the long-term rates for each user requires (in general) a massive utility maximization problem over all the SINRs and BS loads. On the other hand, an actual implementation will likely resort to a simple biasing approach where a BS in tier j is treated as having its SINR multiplied by a factor A_j>=1, which makes it appear more attractive than the heavily-loaded macrocell. This paper bridges the gap between these approaches through several physical relaxations of the network-wide optimal association problem, whose solution is NP hard. We provide a low-complexity distributed algorithm that converges to a near-optimal solution with a theoretical performance guarantee, and we observe that simple per-tier biasing loses surprisingly little, if the bias values A_j are chosen carefully. Numerical results show a large (3.5x) throughput gain for cell-edge users and a 2x rate gain for median users relative to a max received power association.

In a data communication system capable of variable rate transmission, high rate packet data transmission improves utilization of the forward link and decreases the transmission delay. Data transmission on the forward link is time multiplexed and the base station transmits at the highest data rate supported by the forward link at each time slot to one mobile station. The data rate is determined by the largest C/I measurement of the forward link signals as measured at the mobile station. Upon determination of a data packet received in error, the mobile station transmits a NACK message back to the base station. The NACK message results in retransmission of the data packet received in error. The data packets can be transmitted out of sequence by the use of sequence number to identify each data unit within the data packets.

Method and apparatus for high-rate packet data transmission

An IP flow classification system is used in a wireless telecommunications system. The IP flow classification system groups IP flows in a packet-centric wireless point to multi-point telecommunications system. The classification system includes: a wireless base station coupled to a first data network; one or more host workstations coupled to the first data network; one or more subscriber customer premise equipment (CPE) stations in wireless communication with the wireless base station over a shared bandwidth using a packet-centric protocol; and one or more subscriber workstations coupled to each of the subscriber CPE stations over a second network; a resource allocation device optimizes end-user quality of service (QoS) and allocates shared bandwidth among the subscriber CPE stations; an analyzing and scheduling device analyzes and schedules internet protocol (IP) flow over the shared wireless bandwidth. The analyzing device includes the above IP flow classifier that classifies the IP flow. The classifier can include a device for associating a packet of an existing IP flow with the IP flow. The classifier can include a QoS grouping device that groups a packet of a new IP flow into a QoS class grouping. The QoS grouping device can include a determining device that determines and takes into account QoS class groupings for the IP flow. The QoS grouping device can include an optional differentiated services (Diff Serv) device that takes into account an optional Diff Servs field priority marking for the IP flow.

IP-flow classification in a wireless point to multi-point (PTMP) transmission system

A wireless quality of service (QoS) agent for an all-Internet Protocol (IP) network. The QoS agent couples to an all-IP network. The coupling means includes communication means for transfer of information between the agent and a QoS manager of the all-IP network. The agent is also able to seamlessly extend QoS support for multimedia applications from wireline to wireless and control QoS of the multimedia applications sent over wireless connections on the all-IP network.

Method and system for wireless QOS agent for all-IP network

Disclosed is a system for minimizing the effects of faults over an air link of a wireless transmission channel utilizing Transport Control Protocol (TCP). The system includes a TCP-Aware Agent Sublayer (TAS) in a protocol stack, which has a mechanism for caching both TCP packets during forward transmission and acknowledgment (ACK) return packets. The caching mechanism is located near a wireless link of the wireless transmission channel. The system also includes a link monitoring agent coupled to the TAS. The link monitoring agent monitors the condition of the wireless transmission channel for an occurrence of a predefined fault. Once a predefined fault is detected, a system response is implemented based on the type of fault encountered. When the fault is an air link packet loss, an associated packet is immediately retransmitted from the cache, and when the fault is a temporary disconnect, a congestion window of the TCP source is closed.

TCP-aware agent sublayer (TAS) for robust TCP over wireless

We propose a dynamic load balancing scheme for the channel assignment problem in a cellular mobile environment. As an underlying approach, we start with a fixed assignment scheme where each cell is initially allocated a set of channels, each to be assigned on demand to a user in the cell. A cell is classified as ‘hot’, if the degree of coldness of a cell (defined as the ratio of the number of available channels to the total number of channels for that cell), is less than or equal to some threshold value. Otherwise the cell is `cold'. Our load balancing scheme proposes to migrate unused channels from underloaded cells to an overloaded one. This is achieved through borrowing a fixed number of channels from cold cells to a hot one according to a channel borrowing algorithm. A channel assignment strategy is also proposed based on dividing the users in a cell into three broad types—‘new’, ‘departing’, ‘others’—and forming different priority classes of channel demands from these three types of users. Assignment of the local and borrowed channels are performed according to the priority classes. Next, a Markov model for an individual cell is developed, where the state is determined by the number of occupied channels in the cell. The probability for a cell being hot and the call blocking probability in a hot cell are derived, and a method to estimate the value of the threshold is also given. Detailed simulation experiments are carried out in order to evaluate our proposed methodology. The performance of our load balancing scheme is compared with the fixed channel assignment, simple borrowing, and two existing strategies with load balancing (e.g., directed retry and CBWL), and a significant improvement of the system behavior is noted in all cases.

A dynamic load balancing strategy for channel assignment using selective borrowing in cellular mobile environment

A new location update strategy for personal communication services (PCS) networks and its implementation using a genetic algorithm are proposed. Most of the practical cellular mobile systems partition a geographical region into location areas (LAs) and users are made to update on entering a new LA. The main drawback of this scheme is that it does not consider the individual user mobility and call arrival patterns. Combining these factors with the LA‐based approach, we propose an optimal update strategy which determines whether or not a user should update in each LA, and minimizes the average location management cost derived from a user‐specific mobility model and call generation pattern. The location management cost optimization problem is also elegantly solved using a genetic algorithm. Detailed simulation experiments are conducted to capture the effects of mobility and call‐arrival patterns on the location update strategy. The conclusion from this work is that skipping location updates in certain LAs leads to the minimization of the overall location management cost for a user with a specific mobility pattern and even with moderately high call arrival rate.

A selective location update strategy for PCS users

We propose a dynamic load balancing scheme for the channel assignment problem in a cellular mobile environment. As an underlying approach, we start with a fixed assignment scheme where each cell is initially allocated a set of channels, each to be assigned on demand to a user in the cell. A cell is classified as 'hot', if the degree of coldness of a cell (defined as the ratio of the number of available channels to the total number of channels for that cell), is less than or equal to some threshold value. Otherwise the cell is 'cold'. Our load balancing scheme proposes to migrate unused channels from underloaded cells to an overloaded one. This is achieved through borrowing a fixed number of channels from cold cells to a hot one according to a channel borrowing algorithm. A channel assignment strategy is also proposed based on dividing the users in a cell into three broad types – 'new', 'departing', 'others' – and forming different priority classes of channel demands from these three types of users. Assignment of the local and borrowed channels are performed according to the priority classes. Next, a Markov model for an individual cell is developed, where the state is determined by the number of occupied channels in the cell. The probability for a cell being hot and the call blocking probability in a hot cell are derived, and a method to estimate the value of the threshold is also given. Detailed simulation experiments are carried out in order to evaluate our proposed methodology. The performance of our load balancing scheme is compared with the fixed channel assignment, simple borrowing, and two existing strategies with load balancing (e.g., directed retry and CBWL), and a significant improvement of the system behavior is noted in all cases.

Sanjoy Sen

Papers

Method and system for wireless QOS agent for all-IP network

TCP-aware agent sublayer (TAS) for robust TCP over wireless

A dynamic load balancing strategy for channel assignment using selective borrowing in cellular mobile environment

A selective location update strategy for PCS users

A dynamic load balancing strategy for channel assignment using selective borrowing in cellular mobile environment