The Pipeline

Reading below the surface
Using fingerprints to authenticate network users sounds simple enough, but not everyone can use the technology. Skin damage, dirt or dry fingers can make surface-level fingerprint patterns unreadable. In addition, most fingerprint readers use a silicon dye that can cause inaccurate readings.

Silex Technology America said it has solved those problems. Last week the company released a new fingerprint swipe sensor that reads the layer of skin below the epidermis, where fingerprint patterns remain consistent.

The S1, part of Silex’s SX-Biometrics Suite, uses radio frequency technology to penetrate to the inner layer of skin. It gives a more accurate picture of a fingerprint, said Gary Bradt, vice president of Silex America’s biometric business. The S1 works with any desktop or laptop PC. Meanwhile, company officials announced that next week they will begin marketing a new biometric wireless access control device called Bio-NetGuard. It authenticates people who connect to wireless networks via desktop computers, laptop PCs or handheld devices.

Bio-NetGuard uses a Fujitsu MBF200 sensor and complies with 802.11a/b/g and 802.11i wireless standards. It supports as many as 500 users and has a fingerprint match time of 400 milliseconds.

Making data more intelligent
Identifying meaningful patterns in intelligence is a cumbersome and sometimes impossible task, especially if you don’t have the right tools.

Most organizations currently use what William Donahoo, vice president of product management and marketing at Cogito, described as brute-force methods. For the most part, detectives and intelligence analysts pull together multiple systems and sources to sift through massive amounts of unrelated data. They try to make traditional technology do what it was never designed to do, Donahoo said.

A new version of Cogito’s graph-based tool makes it easier for users to view and analyze large sets of data with the goal of detecting patterns and identifying complex relationships.

Cogito Knowledge Center 2.2 models data in a graph overlay, which allows users to discover complicated and distant relationships in seconds, Donahoo said.

The latest version includes new features that allow users to control the process of discovering useful intelligence buried in a dataset without relying on a database administrator for queries and schema changes. Some of those features include faster data importing with support for on-demand requests, scalable graph overlays, custom icon support and label generation for enhanced visualization, and support for unstructured data sources.

Now it is easier for analysts to import data, such as case and event information, and fuse it into a single data structure, Donahoo said. Users can model the information to establish various contextual relationships and views and determine if connections exist between seemingly unrelated data points.

Cogito Knowledge Center runs on Sun Microsystems’ Java and Microsoft’s .NET platforms. It works with service-oriented architecture, Web services application program interfaces and standard data interchange formats.

To expand the reach of Cogito Knowledge Center throughout government agencies, the company is partnering with Booz Allen Hamilton, Donahoo said. The systems integrator has joined the Cogito Solution Provider Program.