COOPERATIVE ROUTING IN WIRELESS NETWORKS A COMPREHENSIVE SURVEY

COOPERATIVE ROUTING IN WIRELESS NETWORKS: A COMPREHENSIVE SURVEY Authors: Fatemeh Mansourkiaie, Mohammed Hossam Ahmed,

Outlines • Summary of the paper • introduction • Background to cooperative routing • taxonomy of routing schemes and the existing cooperative routing • optimality of cooperative routing algorithm • cooperative routing algorithm objectives and aims • Review of centralized and distributed cooperative routing algorithm • Some existing routing algorithms • Challenges and conclusion • Some issues • energy-efficiency, Qo. S parameters, and collision minimization.

Paper summary • Cooperative diversity has gained much interest due to its ability to mitigate multipath fading without using multiple antennas • There has been great amount of research in the utilizing cooperative transmission to improve physical layer performance • Designing and evaluating Cooperative routing algorithm become the interest of many researchers. • What are cooperative routing algorithms? • Routing algorithms that take into consideration the availability of cooperative transmission at the physical layer are known as cooperative routing algorithms. • presents a comprehensive survey of the existing cooperative routing techniques together with the highlights of the performance of each strategy.

Paper summary • Providing classification of existing cooperative protocols techniques and highlight performance of each strategy Outline the main components of the routing protocols. • Describe some of the challenges of each routing protocol •

Cooperative transmission example • Two nodes communicating (src+relay->dest) • Spatial diversity due too independent fading paths • Multiple copies of the same signal arrives at dest • Advantages: Better signal quality • Reduce transmission power • Better coverage • Higher copacity •

Introduction • Cooperative routing allows modes to collaborate with each other to overcome channel fading and • Improve the reliability of wireless networks • • Cooperative communication at the physical layer Cooperative and relaying schemes(amplify-and-forward, decode-and-forward) • Transmission power allocation • Relay selection • • Cooperative MAC protocols: Facilitates cooperative transmission at the physical layer • Modification of DCF • Additional signaling is used HTS(helper ready to send), c. RTs(cooperative request to send; RS(relay start), RA( relay acknowledgement ) and RB(relay broadcasting) •

Introduction(2) • Cross-Layer cooperative Communication: Design that combines network layer and physical layer routings • Enhance the performance of the routing protocol in wireless nets. • • Improves energy efficiency and Qo. S • Path loss is reduced by shortening link length • Less interference due to less power used in transmission

Background • (a and i )are the source and (j and b )are destination • CT we have one or more relay nodes • DT just src and dest

Designing cooperative system Requirements • Making design about cooperative transmission schemes • Relay node selection • Resource allocation • Channel state information(csi) • Cooperative metrics

Cooperative Transmission Scheme • Includes decision whether cooperation is necessary or not • 1) Relaying techniques • Fixed relaying schemes such as DF (decode-and-forward), and AF(Amplify-and-forward) Performance depends on the position of the relay nodes • DF outperforms AF if relay is near transmitter • AF outperforms DF if relay is near receiver node. • DF might offer higher complexity while AF offers data storage problem (analogue signal) • • Adaptive relaying schemes Selection: sender selects the relay based on SNR • Incremental: if destination is unable to detect signal using direct link, relay forwards it •

combining techniques • Maximum Ration Combining(MRC) The received signals from all cooperators are weighted and combined to maximize the instantaneous SNR. • MRC is optimal and maximize the SNR but requires the full knowledge of CSI • • Equal gain combining (EGC) simplified sub-optimal combining technique, where the destination node combines the received copies of the signal by adding them coherently. • Therefore, the required channel information at the receiver node is reduced to the phase information only • • Selection Combining( SC) SC is even simpler and the combiner simply selects the signal with larger SNR. • Although, SC removes the overhead of estimating the channel state information, • its performance is ultimately degraded compared to the MRC and EGC •

Relay Node Selection • It is import for the performance of CR because a good quality relay node yields a higher diversity gain • • Enhance cooperative routing objectives such as energy efficiency, throughput, and packet delivery ratio Optimal number of relay nodes more relay nodes will lead to a higher diversity gain and better performance • But this could cause more interference and requires more resources • • Optimal relay node placement Best relay is the one that enhance performance of cooperative routing • For DF, it is shown that the midpoint between the transmitter-receiver pair in each link •

Resource Allocation in Cooperative Communication • -)Optimal resource allocation: Power, time and bandwidth • Optimization techniques are used to allocate resources • • -)Equal resource allocation Simple but not optimal • Used for some applications •

Channel state information(CSI) • Most of the existing CR utilizes CSI to evaluate the effectiveness of the existing cooperative schemes • In a cooperative link: • the receiver nodes: • • utilize the available CSI for coherent reception, and the transmitter nodes • can utilize the available CSI for power control, signal combining, and relay selection. • To estimate a channel coefficient, a pilot message is usually sent out by the transmitter nodes, and then the receiver nodes exploit the known pilot message to estimate the channel coefficient • Perfect CSI is hard to obtain,

Cooperative Routing metrics • Cost used by routing protocol whether a router should be chosen over another Affect path selection and • Resource consumption • • Metric for cooperative routing protocol is determined for each single link • Common metrics Energy consumption • Packet delivery ratio • Collision probability • • Some cooperative routing protocols depend on multiple metrics based on applications they serve

Taxonomy of Cooperative routing

Optimality • A) optimal Cooperative routing Schemes Jointly use of optimal relay node, optimal power allocation, and path selection • More complex framework is used to achieve the optimal performance • • B)Sub-Optimal cooperative routing schemes • 1)cooperative along non-cooperative path Employing last few nodes as relay node • Using contention among nodes(longer connection time) • Using node location to find the relay set • • 2) cooperative-based Path Node receive msg to form cooperative set • Node is assumed to be closed to the receiver • Neighboring nodes are investigated to find node with minimum cost •

objectives • Saving energy is one of the main objectives • • • Minimizing the total transmission power Sensor networks, ad hoc, and PAN Done through : • • • Energy- efficient cooperative links Energy- efficient relay node Energy- efficient path selection (or mixed of above) • Prolonging the network lifetime • Qo. S • • The time till the first node dies Throughput: depends on SNR, BW, delay, congestion, and collision Packet Delivery Ratio Outage is when destination is unable to detect signal

Decentralization-cont’ • Optimal relay selection, resource allocation and route selection is implemented when all • Centralized • Central controller does : Select relay node(s) and inform transmitter of the relay nodes • Use feedback channels from receiver/relay nodes to collect CSI • Perform global optimization in term of power allocation • Scaling problem when No of users increase(then exchange increases)->complexity • Full knowledge of location of every node in the network to select the cooperative path •

Decentralization • Distributed Each transmitter node responsible to construct rooting, relay selection, storing info about the channel state • Scale better because no dependency on centralized node to choose link or whatsoever • Use some sort of hello message to exchange energy and topology information •

Existing Cooperative Routing • Throughput Optimized Cooperative Routing (TOCR) • Proposed to improve the network throughput by • • Improving the throughput of each link with the help of cooperative relay nodes Based on Adaptive Forwarding Cluster Routing (AFCR): nodes are divided into several 1 -hop cluster • Routing info is exchanged between neighboring nodes and each node in the network has a neighboring table, namely the Cluster Membership (CM) table. • CM table stores cluster-head route • When there are data packets to forward, the node first searches for the destination node in its CM table • Forward it if the node in the cluster, then just forward it, • Else look for the clusterhead •

Existing Cooperative Routing • Throughput Optimized Cooperative Routing (TOCR)(2) • • • Is sub-optimal CR and belongs to Cooperative along non-cooperative Path Each node calculates the throughput of its next hop node and, The throughput gain for potential Cooperative relay node The maximum throughput gain is selected as the best relay node chosen compared to other nodes The complexity of TOCR is related to AFCR

• ㄴ

SUMMARY AND COMPARISON OF COOPERATIVE ROUTING PROTOCOLS

Routing performance Network Lifetime Average Energy per packet • •

challenges • Multiple nodes • Multiple objectives • Node Mobility • Optimal cooperative routing

AN ANALYSIS OF ROUTING PROTOCOL METRICS IN WIRELESS MESH NETWORKS(WMN) Fariborz entezami, Christos Politis

Background • Metric are the key point for routing protocols to select path over path • Hop count for example is the number of wireless links (hops) traversed by a packet between its source and destination • So with the minimum hop count, Path is chosen based on this metric • Packet is forwarded to this path • • Problem with traditional routing makes a design of new metrics that consider wireless environment is essential • Link metrics, Link-Quality metrics, Multichannel metrics

Hop count • Advantages: • • Simple to calculate and has been used for many routing protcols Disadvantages: Does not account for link quality • It is not accurate to estimate the path cost when Hop count is equal for two links • • Solution to routing protocols: Considering more parameters such as interference, path delay, BW • These provide an accurate estimation of the link • These called link-quality metrics(evaluates the quality of each link, cost • (ETX, ETT, and ENT) •

Expect transmission count(ETX) • . ETX estimates the number of transmissions (including retransmissions) required to send a packet over a link. • It is calculated based on packet loss rate(collected from MAC layer) • Suitable for short paths • Does not try to route over congested links • ETX does not consider Based on the probability that packets are successfully transmitted between sender and receiver in bidirectional manner • by each node for each link • • • Overhead from sending Probe packets Broadcast Packet are small and sent in very low rate so it may not reflect the actual big sized data link asymmetry MAC back off Any mechanism to encountering interference

Route ETX = Sum of link ETXs Route ETX Throughput 1 100% 2 50% 3 33% 5 20%

ETX Properties • ETX predicts throughput for short routes (1, 2, and 3 hops) • ETX quantifies loss • ETX quantifies asymmetry • ETX quantifies throughput reduction of longer routes

Big packets

Expected Transmission Time (ETT), Medium Time Metric (MTM), and Weighted Cumulative Expected Transmission Time (WCETT) • ETX does not perform well under certain circumstances (in congested links ) • ETT brings packet size into the path calculation • Suitable for short links because (multiple links have different BW ) • MTM explicitly account for MAC-related overheads in the MTM metric • MTM has been designed for the use of multiple data rate • WCETT first routing protocols metric that consider channel diversity in multi channel networks calculates the Expected Transmission Time (ETT) of each hop and makes the routing decision based on the Cumulative ETT • ETT depends on BW and loss rate • individual link weights are combined into a path metric called Weighted Cumulative ETT (WCETT) •

TRAFFIC AWARE METRICS

distribution based expected transmission count(DBETX) • Enhancement of routing metrics through a more complete view of the physical channel • improves the performance of the network in the presence of varying channels • It exploits this estimation to increase routing efficiency. • The proposed metric is shown to outperform the conventional ETX metric in the presence of fading. • The improvement over ETX increases with the network density because connectivity increases and more routing options become available. Results show a reduction of up to 26% in the Average Number of Transmissions per link and an increase of up to 32% in the end-to-end availability.

distribution based expected transmission count(DBETX)

MULTI CHANNEL METRICS

OLSR metrics • Hop count • • Choose the route with minimum hop count Expected transmission count (ETX) Minimum ETX route is chosen • Selected route have higher throughput as a result • + using Probe message as broadcast • -Probe messages are small • • Minimum loss (ML) selecting the path with the minimum loss(based on ETX) • probability of successful transmission(not both direction as in ETX) •

Multi radio OLSR metrics • Multi radio or multi channel optimized new version of OLSR • Sends data though multiple links to avoid congesting one link and improve throughput • Avoid path with congestion • MR-OLSR uses improved weighted estimate transfer time • Uses also channel allocation strategy and path scheduling algorithms

Evaluation Model

Suggested future work • Using Received signal indication RSSI • Link quality indicator LQI • RF power • Preparing simulation or test bed to study more metrics • Investigating metrics in environment where energy is not major concern