This book is a collection of articles written by leading researchers in information theory stemming from the DIMACS Workshop on Network Information held at Rutgers University (Piscataway, NJ). The articles focus on problems concerning efficient and reliable communication in multi-terminal settings. Information theory has recently attracted renewed attention because of key developments spawning challenging research problems. The material is divided into four parts: 'Information Theory for Sources', which concentrates on network source coding problems; 'Information Theory for Channels', where channels, rather than sources, are central to the problem; 'Information Theory for Sources and Channels', which addresses both source and channel coding; and 'Coding', which deals with more practical issues.Mathematicians using applications such as wireless cellular and LAN data services, ad hoc networks and sensor networks will benefit from the developments outlined in these sections. The book is suitable for graduate students and research mathematicians interested in communications and network information theory.