استشهاد ابن زيدون بأشعار المتنبى في رسالته الجدية التي كتبها في غياهب السجن

This article probes into poetical citation in the historical letter of Ibn-e Zaydun, a renowned Andalusion poet of 11th  century A.D. Ibn-e Zaydun was imprisoned by king of Córdoba, Ibn-e- Jahoor. While in prison, Ibn-e- Zaydun wrote Ibn- e- Jahoor a letter lamenting that he has been thrown into prison for no reason and appealed  for mercy and leniency towards him. The depth of thoughts reflected in the poetic text of Ibn- e- Zaydun`s letter testifying  his command over  poetry. The poet who is  quoted in the letter of Ibn- e- Zaydun is known as Al- Mutanabi. The article examines the parts of the Ibn- e- Zaydun`s letter citing the poetry of Al- Mutanabi in order to make it effective in achieving the objectives of the study.

Scalable Distributed Hash Table Based Routing With Distributed Partition Detection and Mergig I Mobile Ad Hoc Network

Distributed Hash Table (DHT)-based routing in Mobile Ad Hoc Networks (MANETs) completely eliminates flooding at the control plane as well as at the data plane, thus makes the network scalable. In DHT-based MANETs, a logical structured network is built over the ad hoc physical topology in a fully distributed manner and routing is performed using logical identifiers (i.e. transient addresses) rather using IP addresses. This thesis investigates the DHT-based routing paradigm in MANETs and points out that existing state-of-the-art DHT routing protocols assume ideal network environments and, ignore the adversarial environment offered by MANETs. Limited radio range, mobility, lack of infrastructure and decentralized nature introduce frequent and unpredictable changes to network topology in MANETs (i.e. connectivity/dis-connectivity, node(s)/link(s) failure, network partition, frequent merging). The network dynamics severely damage the logical structured network (i.e. the logical space (LS) distributed among the nodes) and completely halt communication. Specifically, existing work fails to address issues such as node(s)/link(s) failure and its impact (i.e. anchor node failure and lookups), network partitioning, lost LS recovery and reusing (i.e. disrupted LS) and merging considerations. Curtailing the information loss due to the network dynamics is imperative for the successful communication in DHT networks. Similarly, the key factor that defines overall routing performance in DHT networks is the successful resolution of lookup requests with minimum possible delay. However, we found that existing DHT protocols suffer from longer delay and, fail to ensure the successful resolution of the lookup requests. Therefore, effective distributed solutions under the scalability constraints are needed to tolerate the faults in the logical network and to provide end-to-end connectivity in such an adversarial environment. It is worthy to mention that the targeted problems are completely unexplored and had never been addressed by the research community. For the first time, we are exploring the problems and providing solutions under the constraints. The first part of our work explores the impact of network dynamics on the intrinsics of DHT routing (i.e. lookup requests and successful resolution with minimum delay). A novel address publication mechanism, also called Anchor Request, is proposed. The mechanisms exploit k-hop topological information to detect critical regions in the network and replicate the index information (stored on the anchor node) across those regions. The considered prefailure measures (i.e critical regions detection and replication) are found to have good side effects on the resolution of lookup requests and delay, despite the failures. Simulation results endorse the significant gains in terms of lookup delay and success ratio. The second part targets the problems of distributed partition detection, unavailability of the anchor node due to partition, lost logical space (LS) recovery and reusing, and merging in DHT networks. We are using the philosophies of detection, replication and recovery to solve the identified problems. The proposed solutions ensure access to the index information despite the network partition. Similarly, prior to the network partition event, LS recovery and reusing is performed, this contributes an evenly distributed and connected logical network. The LS recovery process maintains evenly distributed LSs in each instance of the network after partition. Also, this sets grounds for smooth merging of the disjoint instances. Simulation results confirm the effectiveness of the proposed solutions.