The application of the MINs in Systems-onChip (SoC) and Networks-on-Chip (NoC) is hottest since year 2002. Nevertheless, nobody used them practically for parallel communication. However, to overcome all the previous problems, a new method is proposed that uses Multi-stage Interconnection Network (MIN) to provide intra (global) communication among Application-Specific NoCs in Networks-in-Package (NiP). For this, twin 2 ( ) O n faulttolerant parallel algorithms are proposed. It allows different NoCs to communicate in parallel using either fault-tolerant irregular Penta multi-stage interconnection network (PNN) or fault-tolerant regular Hexa multi-stage interconnection network (HXN). These two are acting as an Interconnects-onChip (IoC) in NiP. Both IoC uses Packet Switching and Wormhole Switching to route packets from one source NoC to destination NoC. The results are compared in terms of packet losses and wormhole switching comes out to be better than packet switching.

Last few years have shown brisk intensification inInternet over the entire globe with heterogeneous networks. Hence the effectiveness of most important & inseparable part, TCP/IP protocol suite, plays a significant role. Wireless links involved in the formation of heterogeneous network is facing serious problems of random BER (Bit Error Rates) which causes random loss. The older TCP (Transmission Control Protocol) was designed for wired links, where the cause of packet loss was due to congestion only. The same protocol policies were incorporated in TCP/IP model with least changes initially. As this was unable to differentiate between losses due to congestion and due to corruption (BERs-random loss), it became the main cause of performance degradation of TCP on wireless links. This paper explores the way to add intelligence in TCP by modifying it to improve data rate in case of random losses which occur on wireless link frequently. Scenarios used to test algorithm are experiencing congestion and corruption loss individually as well as together. Standard BERs are used to measure performance. Modified protocol shows great performance improvement over unmodified in case of random loss as well as shows compatibility with an unmodified protocol in case of congestion loss only.