## Abstract

Interference between nodes directly limits the capacity of mobile ad hoc networks. This paper focuses on spatial interference cancellation with perfect channel state information (CSI), and analyzes the corresponding network capacity. Specifically, by using multiple antennas, zero-forcing beamforming is applied at each receiver for canceling the strongest interferers. Given spatial interference cancellation, the network transmission capacity is analyzed in this paper, which is defined as the maximum transmitting node density under constraints on outage and the signal-to-interference-plus-noise ratio. Assuming that the locations of network nodes are Poisson distributed and spatially i.i.d. Rayleigh fading channels, mathematical tools from stochastic geometry are applied for deriving scaling laws for transmission capacity. Specifically, for a large number of antennas per node, the transmission capacity scales with the number of antennas raised to a fractional power, which depends only on the path-loss exponent. Moreover, for small target outage probability, transmission capacity is proved to increase following a power law, where the exponent is the inverse of the size of antenna array or larger depending on the pass-loss exponent. As shown by simulations, spatial interference cancellation increases transmission capacity by an order of magnitude or more even if only one extra antenna is added to each node.

Original language | English (US) |
---|---|

Title of host publication | 2008 IEEE Global Telecommunications Conference, GLOBECOM 2008 |

Pages | 666-670 |

Number of pages | 5 |

DOIs | |

State | Published - Dec 1 2008 |

Event | 2008 IEEE Global Telecommunications Conference, GLOBECOM 2008 - New Orleans, LA, United States Duration: Nov 30 2008 → Dec 4 2008 |

### Other

Other | 2008 IEEE Global Telecommunications Conference, GLOBECOM 2008 |
---|---|

Country | United States |

City | New Orleans, LA |

Period | 11/30/08 → 12/4/08 |

## ASJC Scopus subject areas

- Electrical and Electronic Engineering